Library

The Library section provides you with a series of tools with which you can add content from multiple sources to your knowledge base. These contents will be used to provide results adjusted to the user's queries. To get started, click 'Add'

Add website

This is a website crawling tool that allows you to extract your content automatically and index it easily.

Automatically

  1. Write the URL of the site to be indexed. It must start with https://

  2. Press 'Scan and add websites'

Advanced Options

There are a number of advanced options you can use to refine the extraction of website content.

  1. Category: important for organizational purposes.

  2. Scan depth: indicates the number of levels that the system should scan (eg 2)

  3. Added Website Limit: Number of addresses to be indexed

  4. Website elements to exclude: You can exclude elements by indicating their selector (eg .banner-header, #footer)

  5. Routes of the website to exclude: the urls that you indicate here will not be scanned.

  6. Automatically keep content synchronized with the source: By activating this box, the content will be updated periodically

Manually

  1. Write a list with URLS of the websites to be indexed. They must start with https://

  2. Press 'Add websites'

Advanced options

  1. Category: important for organizational purposes.

  2. Website elements to exclude: You can exclude elements by indicating their selector (eg .banner-header, #footer)

Add PDF files from your computer

Click on the box or drag a PDF file to upload and process it

Advanced Options

You can add a category to the document and also exclude pages from it, indicating the page numbers separated by commas.

Add custom text

Click on the text box and paste the content that you want to be part of the chatbot. In the advanced options you can add a category.

Improve readability

When we extract content from a website or document, the raw content can be difficult to read. If we select an element added to the library and edit it, we can click on 'Improve document readability'. The extracted text will be adapted and improved to make it more understandable.

Last updated